Examples of Improper Practices:

Forging another’s name or initials
Reporting results without actually analyzing the associated samples  
Signing for review of data when such review was not completed  
Knowingly altering data entries without acceptable justification  
Any action to “cover up” an occurrence from which loss of data and/or revenue can occur
Substituting previously analyzed QC samples  
Spiking a “little extra” to improve recoveries  
Not reporting known instances of improper practices  
Manufacturing and/or distributing controlled substances on DCL property  
Fabrication or falsification of records or data
Improper Peak Integrations
Improper Clock Setting (Time Traveling) or Improper Date/Time Recording
Improper GC/MS Tuning
Discarding points in the initial calibration 
 Performing multiple (more than two) calibrations or QC runs
Using the incorrect (previous) initial calibration
Improper Alteration of Analytical Conditions
Discarding points from a MDL study without use of established procedures and required documentation
Misrepresentation of QC Samples and Spikes
Adding surrogates after sample extraction
Reporting post-digested spikes or duplicates as pre-digested spikes or duplicates.
Not preparing or analyzing method blanks and laboratory control samples (LCSs) the same way that samples are prepared or analyzed in order to make it appear that method blank or LCS results are acceptable when, in fact, they are not.
Raw Data File Substitution
Overdilution of Samples or Misrepresentation of Detection Limits
Deletion of Noncompliant Data
Concealment of a Known Problem 
Unwarranted Manipulation of Computer Software

 

Problem

Unacceptable Solution

Acceptable Solution

Lack of time or resources to perform testing

Making up Data or Other Information – Creating data for an analysis that was not performed or creating information that is not true.

Analytical results for all samples and quality control (QC) must be based on actual analyses performed. Documented data must match actual data. Sampling information must be based on actual sampling events.

Hold time near or past

Improper Clock Setting (Time Traveling) or Improper Date/Time Recording – Resetting the internal clock on an instrument to make it appear that a sample(s) was analyzed within a specified hold time when, in fact, it was not. Changing the actual time or recording a false time to make it appear that hold times were met, or changing the times for sample collection, extractions, or other steps to make it appear that they were performed at the correct time when, in fact, they were not.

The recorded date and time of collection, preparation, or analysis must match the actual date and time that the action was performed. Documented dates and times must repre­sent actual dates and times. Samples exceeding hold times must be reported as such; a case narrative is recommended.

DFTPP or BFB not meeting acceptance criteria

Improper GC/MS Tuning – Artificially manipulating GC/MS tuning data to produce an ion abundance result that appears to meet specific QC criteria when, in fact, the criteria were not met.

GC/MS tuning data must be generated and reported according to proper techniques without manipulation of the peak or mass spectrum. Preventive/corrective action must be taken concerning data not meeting required criteria.

Calibration or QC data not meeting acceptance criteria

Improper Peak Integration (Peak Shaving or Enhancing) – Artificially subtracting or adding peak area to produce an erroneous area that forces data to meet specific QC criteria when, in fact, the criteria were not met.

Instrument peaks must be consistently integrated and reported according to proper techniques, generally baseline-to-baseline, valley-to-valley, or a combination of the two. Peak area cannot be subtracted or added to force data to meet specified criteria. Preventive corrective action must be taken on instrument data not meeting required criteria.

Calibration or QC data not meeting acceptance criteria

Improper Calibration/QC Analysis –
a)  Performing multiple (more than two) calibrations or QC runs (including calibration verifications, LCSs, spikes, duplicates, and blanks) until one analysis barely meets criteria, rather than taking needed preventive/corrective action after the second failed analysis, and not document­ing or retaining data for the other unacceptable data.

b) Using the incorrect (previous) initial calibration to make calibration verification data appear to be acceptable when, in fact, they were not acceptable when compared to the correct initial calibration.

c)  Discarding points in the initial calibration to force the calibration to meet acceptance criteria.

d)  Discarding points from an MDL study to force the calculated MDL to be higher or lower than the actual value.

a) All calibration and QC data associated with sample analyses must be documented. Preventive/corrective action must be taken and documented if calibration and/or other QC criteria are not met.

b) Acceptance of calibration verification data must be based on the correct initial calibration.

c) Calibration points can only be rejected for inclusion in the calibration curve if a known error was made or if a statistical evaluation indicates that a point can be discarded. When multiple target analytes are included in each calibra­tion standard, it might be necessary to discard selected upper or lower points for individual target analytes. Points can be discarded at the upper end of the curve if the linear range of the detector has been exceeded. For these cases, dilute samples that exceed the highest point of the calibra­tion curve. Points can be discarded at the lower end of the curve if the detector is not producing a response. For these cases, the laboratory reporting limit must be adjusted accordingly.

d) Data points for MDL studies can only be rejected for inclusion in the MDL calculation if a known error was made or if a statistical evaluation indicates that a point can be discarded.
QC samples or spikes not meeting acceptance criteria

Misrepresentation of QC Samples and Spikes –Misrepresenting QC samples or spikes as being digested or extracted when, in fact, they were not actually digested or extracted. For example:

a) Adding surrogates after sample extraction rather than prior to sample extraction.

b) Reporting post-digested spikes or duplicates as pre-digested spikes or duplicates.

c) Not preparing or analyzing method blanks and laboratory control samples (LCSs) the same way that samples are prepared or analyzed in order to make it appear that method blank or LCS results are acceptable when, in fact, they are not.

 

QC samples and spikes must be prepared, analyzed, and reported according to appropriate procedures.

a) Surrogates must be added prior to sample extraction.

b) Post-digestion spikes and duplicates must be reported as post-digested and must not be  misrepresented as pre-digestion spikes and duplicates.

c) Method blanks and LCSs must be prepared and analyzed the same way that samples are prepared and analyzed. Any QC results outside acceptance criteria must be reported as such; a case narrative is recommended.

 
Calibration or QC data not meeting acceptance criteria File Substitution – Substituting previously generated files (runs) for a noncompliant calibration or QC run to make it appear that an acceptable run was performed when, in fact, it was not. Data must be generated and reported for actual analyses performed. Reported dates and times for all analyses must match actual dates and times. Substitution of files is not permitted.
Calibration or QC data not meeting acceptance criteria Unwarranted Manipulation of Computer Software – Unwarranted manipulation of computer software to force calibration or QC data to meet criteria, and removing computer operational codes, such as “M” flag. Computer manipulation is allowed only for warranted reasons, and any manipulation should be minimal and traceable. Removal of computer operational codes is not permitted.
Analytical conditions for standards do not work for samples Improper Alteration of Analytical Conditions –Improperly altering analytical conditions, such as changing the instrument conditions for sample analyses from those used for standard analyses. Using different procedures to process sample data than those used for standards. All sample analyses must be performed under the same conditions as those used for standard analyses. Any alteration of analytical conditions must be allowable under the method requirements. All sample data must be proc­essed by the same procedures as those used for processing standard data. Any discrepancies must be documented.
Sample not analyzed at appropriate level or not reported at correct detection limit Overdilution of Samples or Misrepresentation of Detection Limits – Intentionally diluting a sample to such an extent that no analytes (target or non-target) are detected without justification as to why the high dilution was made. Reporting a detection limit that does not represent the sample analysis (e.g., not including dilution factor in sample detection limit) Dilutions must be made on a reasonable basis, such as high concentrations of target or non-target analytes, matrix interferences, oily samples, and other components in the sample that could harm the instrument. Include details concerning the reason for the dilution in a case narrative. Sample detection or reporting limits must include dilution factors.
Noncompliant data Deletion of Noncompliant Data – Intentional deletion or non-recording of noncompliant data to conceal the fact that analyses were noncompliant All data associated with sample collection and analysis, including any out-of-control events or noncompliant data, must be documented and retained. Preventive/corrective action must be taken and documented for any noncompliant data.
Undesirable situation with analysis or sample; knowledge of unethical conduct Concealment of a Known Problem – Concealing a known analytical or sample problem from laboratory manage­ment and/or client. Concealing a known unethical behav­ior or actions from laboratory or corporate management Any knowledge of analytical or sample problems must be communicated to laboratory management and the client. Any knowledge of unethical behavior or actions must be fully communicated to laboratory or corporate management.